1

I have to create some species distribution maps, but I have very little experience with ArcGIS unfortunately.

I have two different kinds of excel sheets: 1. A version with many thousand X and Y coordinates 2. A version that is identical like 1), but has an additional third column saying how many species occur at every X/Y coordinate.

I was told the first one would be the right data sheet, but don't know it that is true.

Anyway, what I need to do, is to somehow import these data into ArcMap and create a species richness map, showing where species occur and with a color-code, how many there are. That is it. This does not sound too complicated, but I can't get it to work.

Could somebody help me and tell me what to do?

  • Welcome to GIS SE. What is your ArcGIS version ? – radouxju Mar 14 '15 at 19:31
  • Sorry, I missed that question. My version is 10.2.2. It should be fairly new – Sara Mar 15 '15 at 9:59
2

I can't be sure from your description but it reads as though the second sheet is identical to the first one, the only difference being the additional description column. So I'll work on that basis.

It reads as though some or all of your x and y values may be text. So the first thing I would do is add another two columns (say, 'X_new' and 'Y_new') and, assuming your x and y coordinates are in columns A and B, write a formula in column D which is =A2*1 - this will force Excel to read the value in A2 as a number. Copy the formula across one column and down so that all your original x and y coordinates have been converted to numerical values.

A useful tip at this point is to rename your Excel tab something pertinent to the subject to be displayed (say, 'Species') - when you have a lot of Excel tables in your Layers menu, all called 'Sheet1', you'll quickly find out what I mean.

Save your Excel spreadsheet as a .csv file and import into ArcGIS. Right click on 'Species' and select 'Display XY Data'. Select 'X_new' as your eastings and 'Y_new' as your northings (I think that's correct, I'm doing all this from memory). You will need to specify a coordinate system - you don't specify the country the data is in, but if it's Britain you'll need to find OSGB_1936 and select that as your coordinate system.

I believe the next stage will involve applying Symbology, but before we go down that route I want to make sure you've successfully imported and plotted the correct data set.

  • Thank you for your kind reply!! From what I read in help sections etc., I can either import the datasheet via File->AddData->Add XY Data or per tool called "Make XY event layer". I have no idea what the difference is. In either way it looks weird. I have no idea what coordinate system I have to specify. The coordinates should basically cover every ocean. But right now it only looks like a part of it. – Sara Mar 14 '15 at 23:01
  • What would I have to choose in the symbology to get different colors for different amount of species per coordinate? I tried around but nothing seems to work. Thanks Bruce! – Sara Mar 14 '15 at 23:03
2

I can't tell you how to do it in ArcGIS, but with QGIS (free) it's pretty easy and straight forward.
Note: Stick with the 2.6.1 version, as 2.8 is still a bit buggy.

As i do not have your data, i downloaded some maritime census data from OBIS.

1) Open QGIS and install the Heatmap and OpenLayers Plugins
SCR01.PNG

2) Import Web>OpenLayers>GMaps>Satellite as a background reference to your project.
SCR02.PNG

3) Import your datatable to your project, define your X/Y columns, create a spatial index and set the CRS (most likely WGS84/EPSG:4326).
SCR03.PNG

Your project should now look smth. like this:
SCR04.PNG

Looks a bit messy with all that points, so we "convert" them to a heatmap

4) Open the Raster>Heatmap>Heatmap dialog, select your point data as "input" and name a path for your output-file.
The most important setting is the

Radius
and to an extend rows/columns.
Rows/Columns define the resolution of the output-raster, Radius controlls the smoothing of the hotspots .. this will need some try&error.
SCR05.PNG

Depending on your parameter, the calculation may take some time, but eventually leaves you with something like this:
SCR06.PNG

5) Now we style the output:
double click on the heatmap in your layer-list to open the properties dialog (1)
set the render-type to "pseudocolor" and choose colors of your liking (2)
set the accuracy to "actual" and click "load" to load the values(3)
classify the values(4)
double click on the "0" value and set the opacity to full transparent (5)
SCR07.PNG

Your output will look something like this:
SCR08.PNG

zoomed in:
SCR09.PNG

with some other "Radius" setting in #4:
SCR10.PNG

Hope that helps.

  • Thank you very much for your reply and your effort !! After hours of try and error I gave up and asked my supervisor. A first mistake was my datasheet. I have new one now, with 3 columns: Lon, Lat and number of species. So obviously what I need to do is to import this and, with graduated colors, show the species richness. But it just looks weird. Maybe there is another mistake in the data. You think you could try with my data quickly to check if there is a problem? Thank you!!! – Sara Mar 15 '15 at 11:13
  • Sure, just upload it somewhere. – ymirsson Mar 15 '15 at 12:06
  • Thank you ymirsson for your kind offer but I found my mistake. There was really an issue with the Excel-sheet, I dont know why. But I got it fixed. I opened another question where I try to improve the resulting map. Maybe you have a tip. Thank you!! – Sara Mar 15 '15 at 14:50
  • @ymirsson in arcgis you can use the 'point density' or 'kernel density' tool, which is similar to Qgis 'Heatmap' plugin. However, in arcgis kernel possibilities are limited to the default, which is not specified clearly, as far as I know. – dof1985 Mar 15 '15 at 15:46

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.